Question Answering on Bio poison @ Position 1, k=10 (test)
79.3Rob. LLM-J ScoreMIS
Evaluation Results
| Method | Links | |
|---|---|---|
| MISModel=GPT-4o-mini, k=102025.09 | 79.3 | |
| Vanilla RAGModel=GPT-4o-mini, k=102025.09 | 75.2 | |
| InstructRAGModel=GPT-4o-mini, k=102025.09 | 69.4 | |
| MISModel=Llama3.2-3B, k=102025.09 | 68.7 | |
| MISModel=Mistral-7B, k=102025.09 | 67.2 | |
| Vanilla RAGModel=Mistral-7B, k=102025.09 | 64.1 | |
| Vanilla RAGModel=Llama3.2-3B, k=102025.09 | 63.6 | |
| InstructRAGModel=Mistral-7B, k=102025.09 | 62.1 | |
| AstuteRAGModel=GPT-4o-mini, k=102025.09 | 62 | |
| RobustRAGModel=GPT-4o-mini, k=102025.09 | 60.9 | |
| RobustRAGModel=Mistral-7B, k=102025.09 | 55.5 | |
| InstructRAGModel=Llama3.2-3B, k=102025.09 | 54 | |
| RobustRAGModel=Llama3.2-3B, k=102025.09 | 52.9 | |
| AstuteRAGModel=Mistral-7B, k=102025.09 | 48.9 | |
| AstuteRAGModel=Llama3.2-3B, k=102025.09 | 35.8 |